This article appeared in a journal published by Elsevier. The attached
copy is furnished to the author for internal non-commercial research
and education use, including for instruction at the authors institution
and sharing with colleagues.
Other uses, including reproduction and distribution, or selling or
licensing copies, or posting to personal, institutional or third party
websites are prohibited.
In most cases authors are permitted to post their version of the
article (e.g. in Word or Tex form) to their personal website or
institutional repository. Authors requiring further information
regarding Elsevier’s archiving and manuscript policies are
encouraged to visit:
http://www.elsevier.com/copyright
Author's personal copy
Journal of Great Lakes Research 36 (2010) 30–41
Contents lists available at ScienceDirect
Journal of Great Lakes Research
j o u r n a l h o m e p a g e : w w w. e l s e v i e r. c o m / l o c a t e / j g l r
Approaching storm: Disappearing winter bloom in Lake Michigan
W. Charles Kerfoot a,⁎, Foad Yousef a, Sarah A. Green b, Judith W. Budd c,
David J. Schwab d, Henry A. Vanderploeg d
a
Lake Superior Ecosystem Research Center and Department of Biological Sciences, Michigan Technological University, Houghton, MI 49931, USA
Department of Chemistry, Michigan Technological University, Houghton, MI 49931, USA
Department of Geological Engineering and Sciences, Michigan Technological University, Houghton, MI 49931, USA
d
NOAA Great Lakes Environmental Research Laboratory, 2205 Commonwealth Blvd., Ann Arbor, MI 48105, USA
b
c
a r t i c l e
i n f o
Article history:
Received 1 January 2010
Accepted 15 March 2010
Communicated by G. Fahnenstiel
Index words:
Winter chlorophyll
SeaWiFS
Lake Michigan
Quagga
a b s t r a c t
Between 1990 and 2001, late-winter phytoplankton blooms were common in parts of the lower Great Lakes
(southern Lake Michigan, Saginaw Bay and southern Lake Huron, and western Lake Erie), providing
resources for over-wintering zooplankton. In Lake Michigan up to 2001, detailed remote sensing and ship
studies documented well-developed late-winter blooms in the southern gyre (circular bloom termed the
‘doughnut’). However, from 2001 to 2008, the winter blooms in Lake Michigan also supported early season
veliger larvae from the introduced, cold-water adapted “profunda” morph of quagga mussels (Dreissena
rostriformis bugensis). Remote sensing and ship studies revealed that settled mussels caused an extraordinary
increase in water transparency and a simultaneous decrease of Chl a in the late-winter bloom. Before quagga
mussels in 2001, water transparency was 74–85% at deep-water sites, whereas it increased progressively to
89% by 2006 and 94–96% by 2008. Chlorophyll a concentrations in the gyre rings were 1.1–2.6 µg/L in 2001,
declining to 0.5–1.7 µg/L by 2006 and 0.4–1.5 µg/L by 2008. The reduction of Chl a in the winter bloom rings
from 2001 to 2008 was 56–78% for the western limb and 74–75% for the eastern limb. Zooplankton species
abundance, composition and abundance also changed, as cyclopoid copepods became very scarce and overwintering omnivorous calanoid copepods declined. Reduction in late-winter phytoplankton and zooplankton
poses a serious threat to open-water food webs.
© 2010 Elsevier B.V. All rights reserved.
Introduction
Prior to the development of remote sensing, details of winter
circulation and production in the Great Lakes were poorly known, in
part because of hazardous ship conditions (Eadie et al., 1996; Kerfoot
et al., 2004). On the National Science Foundation (NSF) and National
Oceanic and Atmospheric Administration (NOAA) Episodic EventsGreat Lakes Experiment (EEGLE) project, we discovered a late-winter
algal bloom in southern Lake Michigan (Chl a ‘doughnut’ ring; Budd
et al., 2002; Kerfoot et al., 2008). Contrary to expectations of fairly
uniform open waters from February to April, sea-viewing wide fieldof-view sensor (SeaWiFS) and moderate-resolution imaging spectroradiometer (MODIS) imagery uncovered spatially complex resuspended sediment, Chl a, and CDOM patterns in offshore waters.
Apparently, phosphorus-rich coastal river waters and nearshore
sediments (Biddanda and Cotner, 2002) were captured by intensified
winter currents, entrained along gyre convergence zones and moved
⁎ Corresponding author.
E-mail addresses: wkerfoot@mtu.edu (W.C. Kerfoot), fyousef@mtu.edu (F. Yousef),
sgreen@mtu.edu (S.A. Green), jwbudd@mtu.edu (J.W. Budd), david.schwab@noaa.gov
(D.J. Schwab), henry.vanderploeg@noaa.gov (H.A. Vanderploeg).
0380-1330/$ – see front matter © 2010 Elsevier B.V. All rights reserved.
doi:10.1016/j.jglr.2010.04.010
into deeper waters, stimulating a ring of offshore production (Kerfoot
et al., 2004, 2008). Cross-lake surveys (April 2001, 2006) with two
separate profiling instruments revealed columnar patterns consistent
with a spatially complex, rotating gyre structure. Optical plankton
counter (OPC) transects and zooplankton net tows indicated that
spatial heterogeneity extended to higher levels of food webs.
Here we note that the late-winter blooms were not peculiar to the
southern basin of Lake Michigan, but also occurred regularly in other
Great Lake waters, where the geometry of bloom patterns differed due
to bathymetry, circulation and lake orientation (Beletsky et al., 1999;
Beletsky and Schwab, 2001). In Lake Michigan, there were indications
from remote sensing and ship-board studies that the winter ‘lateral
nutrient displacement’ phenomenon had been present for at least
25 years. Up to 2001, the late-winter bloom was probably an
important feature that benefited predominately cold, deep-water
native taxa (phytoplankton and zooplankton), and that normally
supported over-wintering strategies.
The degree that global climate change is involved in accentuating
late-winter blooms is unknown, yet global climate change appears to
be reducing ice-cover (Austin and Colman, 2007) and increasing the
frequency and intensity of winter storms (Schwab et al., 2006), thus
promoting the observed lateral displacement of limiting nutrients by
capturing river discharges, resuspended sediments, and redirecting
Author's personal copy
W.C. Kerfoot et al. / Journal of Great Lakes Research 36 (2010) 30–41
31
coastal algae (Kerfoot et al., 2008). Usually these processes would
have enhanced late-winter productivity from 2001 to 2008 in Lake
Michigan. However, this has not been the case as cold-adapted veliger
larvae of the introduced quagga mussel (Dreissena rostriformis
bugensis) have also exploited the late-winter productivity pulses.
Water column filtration from settled adults is now seriously
compromising anticipated late-winter bloom patterns.
It is truly remarkable that one or two species of mussels can so
transform the entire ecosystem. During the EEGLE study, zebra mussel
effects were confined largely to coastal waters off Chicago (Nalepa
et al., 1998; Vanderploeg et al., 2007). In recent years, a ‘shallowwater morph’ of the quagga mussel (Dreissena rostriformis bugensis)
has begun to replace zebra mussels in shallow areas of the Great Lakes
(Nalepa et al., 2009). In deeper (N30 m), cooler waters, another form
of D. rostriformis bugensis (the ‘profunda morph’) has increased
tremendously (Nalepa et al., 2009, 2010). Unlike the zebra mussel, the
‘profunda morph’ is adapted for life on soft sediments (has elongated
incurrent siphon and lies on the bottom on one of its valves without
attachment with byssal threads). The life history cycle for quagga
mussels begins somewhat earlier than zebra mussels and under lower
temperatures (e.g., Claxton and Mackie, 1998). Quagga mussel
spawning occurs in early winter, late spring, late summer, and fall
(Roe and MacIsaac, 1997; Claxton and Mackie, 1998; Nalepa et al.,
2010). The embryos first develop into swimming larvae. In 2–9 days
they develop intestines and a feeding and swimming organ known as
the velum. Once the velum appears, the larvae are termed veligers. In
the veliger stage they develop D-shaped (straight-hinged) shells
about 70–100 µm long (Sprung, 1993). The settlement of mussels to
the bottom varies as a function of temperature and food concentration. At temperatures of 22–26 °C, settlement can occur after 32 d
(Wright et al., 1996), yet it can take as long as several months at low
temperatures and food concentrations.
Moderately high densities of quagga veligers feed on the latewinter bloom pulse, and are dispersed by strong currents associated
with winter storms (Kerfoot et al., 2008). This soft-sediment species
can settle in both shallow and deep waters, reaching 5–15 thousand
individuals/m2. In Lake Michigan, adult quagga mussels increased
dramatically in density between 2001 and 2006, spreading below
100 m depth (Nalepa et al., 2009; 2010). At high densities, adult
quagga mussels are filtering much of the overlying water column
along the coastal shelf (Vanderploeg et al., 2010), seriously reducing
the typical spring bloom.
Here we first discuss long-term evidence for late-winter blooms in
the Great Lakes between 1998 and 2001, then focus in on Lake Michigan.
Remote sensing and coordinated ship studies document how the adult
mussel filtration effects are altering late-winter waters and the circular
late-winter phytoplankton bloom (‘doughnut’). Our observations
1) chronicle the development of a late-winter ‘bottleneck’ in Lake
Michigan and 2) raise grave concerns about the consequences of future
productivity losses on winter pelagic foodwebs.
spatial extent and duration varies from year to year (Ji et al., 2002;
Kerfoot et al., 2004, 2008; Stroud et al., 2009). Since the sensor swaths
included the entire Laurentian Great Lakes region, SeaWiFS image
processing (1998–2002) was extended to the other lakes.
During the early EEGLE SeaWiFS processing, we became aware of the
‘doughnut-shaped’ bloom pattern of Chl a in offshore Lake Michigan
waters (Budd et al., 2002). The SeaWiFS sensor was located on the
PorbView-2-satellite, which imaged the Great Lakes between 17:30 and
19:30 h coordinated universal time (Gregg, 1992; approximately the
same as Greenwich mean time). The instrument had a scan coverage of
2800 km, a nadir resolution of 1.1 km2, and contained a passive, eightband multispectral scanner. The scanner picked up reflectance in six
visible bands (412 nm, Gelbstoffe; 443 nm, chlorophyll; 490 nm,
pigment; 510 nm, chlorophyll; 555 nm, sediments and pigments; and
670 nm, atmospheric correction) used for estimating pigment and total
suspended material (TSM) concentrations, and two near-infrared bands
(765 nm and 865 nm) used primarily for atmospheric corrections
(McClain et al., 1998).
Time series SeaWiFS imagery was processed using SeaWiFSMAP
image processing software (Warrington, 2001) and a modified IDL/
SeaDAS (http://seadas.gsfc.nasa.gov/seadas/) code that included the
1998 sensor calibration (McClain et al., 1998) and the coastal
atmospheric correction scheme (Stumpf et al., 2000, 2003; Budd
and Warrington, 2004). SeaWiFS Chl a (OC2) maps were derived from
empirical, band-ratioing algorithms, OC2v4 (O'Reilly et al., 2000a) and
OC4v4 (O'Reilly et al., 2000b). The OC2v4 Chl a algorithm used a ratio
of SeaWiFS bands 3 and 5 (O'Reilly et al., 1998; Budd and Warrington,
2004). See Kerfoot et al. (2008) for OC2, OC2v4, and OC4v4 equations
and more details on procedures.
To check for long-term spatial trends in the southern Great Lakes,
we utilized OC2 images and scored the incidence of late-winter bloom
patterns between 1998 and 2002. Spectral reflectance (RRS) at 555 nm
was used to estimate total suspended matter (TSM; McClain et al.,
1998). The ability to observe late-winter blooms is greatly hindered
by cloud cover (Stroud et al., 2009), but the observations provided
strong evidence for past widespread incidence. To provide a
quantitative temporal record of Chl a concentration decline within
the Lake Michigan ‘doughnut’, we superimposed the southern ship
transect on SeaWiFS images between 1998 and 2008 and chose Chl a
pixel values from the inner and outer rims of the eastern and western
‘doughnut’ bands. We then regressed the mean values against time,
over the complete 1998–2008 interval.
Concerns about coastal suspended matter (TSM) and colored
dissolved organic matter (CDOM) influencing OC2 interpretations are
discussed in Stumpf (2001), Stumpf et al. (2003) and Budd and
Warrington (2004). CDOM interference was severe for Lake Superior
OC2 Chl a estimates, but not for Lake Michigan Chl a estimates (Budd
and Warrington, 2004). The entire set of processed images bearing on
pre-quagga conditions from 1998 to 2002 is posted at: http://www.
geo.mtu.edu/great_lakes/lakersi/cgi-bin/seawifs.cgi.
Methods and materials
Phytoplankton distribution and Chl a concentrations
Remote sensing (AVHRR, SeaWiFS, MODIS) imagery
Several cruises were used to validate SeaWiFS imagery and to
determine water column variables. Ship-based Chl a and TSM samples
were collected from January to November 1998–2000 along five
transects off Racine, Chicago, Gary, St. Joseph, Muskegon, and at two
central stations (Fig. 1) and used in initial Chl a on CSAT and R555
(SeaWiFS) regressions (Table 1, regression 1). These initial EEGLE
determinations were supplemented in later years by additional cruise
measurements along the Lake Michigan southern transect. Field and
laboratory calibration curves utilized a pure Chl a standard (Anacystis
nidulans, Sigma C6144). Between 1 and 3 L of lake water from 0, 10,
20, to 25 m depths were filtered (preweighted Whatman GF/F 55 mm
filters) at 30–40 stations along the 62-station southern transect. Both
field and laboratory values for Chl a were run on a Turner TD-700
As part of the NSF/NOAA EEGLE Project in Lake Michigan, we
processed Advanced Very High Resolution Radiometer (AVHRR 1992–
2002) and sea-viewing-wide-field-of-view sensor (SeaWiFS 1997–
2002) imagery (Warrington, 2001; Kerfoot et al., 2004). Since
becoming routinely available in 1992, AVHRR (NOAA TIROS-N series,
NOAA-10 and NOAA-11) imagery has confirmed near-coastal and
offshore sediment plumes (channel 1–3 visible bands) and
corresponding temperature structure (3–5 bands) every year during
late winter (March–April) in Lake Michigan (Leshkevich et al., 1993;
Warrington, 2001; Budd and Warrington, 2004). SeaWiFS and MODIS
imagery has also verified late-winter offshore structures, although the
Author's personal copy
32
W.C. Kerfoot et al. / Journal of Great Lakes Research 36 (2010) 30–41
measures were checked by running a dilution series of previously
calibrated algal cultures (Chlamydomonas) through the fluorescence
sensors.
Near-surface Chl a values were used to validate SeaWiFS OC2v4
reflectance readings. Because field surface samples often showed
daylight Chl a quenching (Kerfoot et al., 2008), which influenced CTD
fluorometry readings, 10–20 m Chl a determinations were favored for
validation regressions over surface values. However, winter Chl a values
for 10–25 m readings were spatially similar and highly correlated. For
example, on the April 2008 cruises, correlations between 10 m, 20 m,
and 25 m Chl a values along the ship sampling transect were very high
(10 m vs 20 m r2 = 0.976; 10 m vs 25 m r2 = 0.974; 20 m on 25 m
r2 = 0.990; N = 74), reflecting winter deep-water vertical mixing.
A previously published relationship between SeaWiFS CSAT and
R555 estimates and in situ TSM concentrations (regression 6, Table 1)
was used to look at long-term trends. The ground-truth effort
provided an empirical nonlinear relationship between SeaWiFS RRS
(555) and ship-based TSM measurements over the observed range of
1–12 mg/L. The logarithmic fit was heuristic for it covered saturating
values in nearshore turbidity plumes, while revealing subtle patterns
for low-concentration levels in offshore waters. Further discussion of
validations used with SeaWiFS imagery during the EEGLE project can
be found in Bergmann et al. (2004) and Lohrenz et al. (2008).
Bergmann et al. (2004) discuss difficulties in OC2 Chl a determinations relating to marine versus freshwater phytoplankton species
differences. Additional water column calibrations between CTD
fluorometry and Chl a are discussed in Vanderploeg et al. (2007)
and Lohrenz et al. (2008).
Fig. 1. Southern Lake Michigan, indicating transect stations for 1998–2000 ship-board
TSM:Chl a validation studies (connected squares off various cities) and the April 2001,
2006, and 2008 Seabird CTD/BAT southern transects (62 stations, solid line from South
Haven to Waukegan). Gyre circulation of waters after a northwest wind is indicated by
arrows, with the position of the convergence zone off Grand Haven/Muskegon noted.
Blue area is region of intensified monitoring during the EEGLE project and region of
greatest phosphate loading from river discharges.
Fluorometer using the acid (phaeophytin) correction method (Wetzel
and Likens, 2000) on 90% acetone extracts. The 2006–2008 cruises
featured a combination of ship-board Chl a determinations supplemented by later laboratory determinations on additional filtered
samples. The field and laboratory filtered samples were used routinely
to calibrate CTD and Acrobat (Wet Lab ECO-AFL) fluorescence values
(Table 1, regressions 3–6). In addition, CTD and Acrobat Chl a
Table 1
Validation and empirical relationships for southern Lake Michigan: coefficients of linear
(Y = bX − a) regressions relating ship-sampled Chl a (μg/L) to CSAT Chl a (μg/L) or CTD
fluorescence (FL). In the first regression, Chl a measurements taken along numerous
transect stations at 5–10 m depth from 22 Mar 98–11 Sep 99 are regressed on SeaWiFS
CSAT Chl a values, whereas the second regression utilizes Chl a measurements from the
southern transect, 21–23 April 08. In the third to fifth regressions (2001, 2006, 2008),
Chl a from southern transect stations is regressed on Seabird CTD fluorescence (FL)
values. The lowest, log-linear regression (equation 6) relates SeaWiFS R555 values to
ship-sampled TSM (mg/L; total suspended matter). TSM measurements for R555
validation were taken Jan–Mar 1998 and Feb–Apr 1999 from 5 to 10 m depths at
various coastal transect stations plotted in Fig. 1. All regressions are significant at the
p b 0.001 level.
Linear regression
n
b ± SE
a ± SE
r2
1.
2.
3.
4.
5.
18
51
36
31
52
1.079 ± 0.102
1.209 ± 0.058
23.433 ± 1.81
21.331 ± 2.54
24.274 ± 1.29
− 0.087 ± 0.171
− 0.022 ± 0.082
− 0.501 ± 0.097
− 0.349 ± 0.196
− 0.398 ± 0.106
0.874
0.900
0.836
0.708
0.872
32
5.975 ± 0.387
2.005 ± 0.15
0.891
Chl
Chl
Chl
Chl
Chl
a on
a on
a on
a on
a on
CSAT 1998–1999
CSAT 2008
FL(CTD) 2001
FL(CTD) 2006
FL(CTD) 2008
Log-linear regression
6. R555 on log(TSM)
Chlorophyll a and zooplankton along southern Lake Michigan transect:
Developing impact of quagga mussels
Four cruises (April 2001, 2006–2008) aboard the RV Laurentian
compared the spatial nature of the late-winter bloom (‘doughnut’) in
southern Lake Michigan, three along the same latitudinal transect
(2001, 2006, 2008). All four cruises originated from the eastern port of
Muskegon, Michigan. The three E–W cruises (Fig. 1) traversed along a
mean latitude of 42.372° from South Haven to Waukegon and back,
cutting across the center of the circular algal bloom on 19–21 April
2001 and along its southern ring on 12–15 April 2006 and 23–25 April
2008. Over the 3–4 day cruises, Seabird SBE 911plus conductivitytemperature-depth (CTD) full vertical profiles were taken from
around 64 stations at 1.85-km (1 nautical mile) intervals (longitude
87.745°W to 86.320°W). The Seabird CTD instrument package
recorded depth, temperature, Chl a fluorescence, photosynthetically
available radiation (PAR), and water transmissivity. The device was
equipped with a SeaTech transmissometer, Biospherical Instruments
scalar PAR sensor (400 to 700 nm), and Wet Lab ECO-AFL fluorometer.
Contour plots were generated from the vertical CTD casts using
Matlab (Mathworks). The CTD data were binned at 0.5-m depth
intervals from casts every 1.85 km for a total of ca. 10,400 points for
each parameter. Data were interpolated in the horizontal dimension
using the cubic polynomial function to provide ca. 32,000 points for
contouring the cross-sections (Kerfoot et al., 2008). On April 2–3,
2007, we traversed diagonally towards the ‘doughnut hole’ from
Muskegon harbor, taking CTD vertical profiles, but were forced off the
lake by an approaching storm. On April 25, 2008, we traversed
diagonally back to Muskegon along a similar transect, and then along
the nearshore coastline.
In addition to CTD vertical casts, we towed a BAT (Sea Sciences
Acrobat) in tow-yo undulating fashion (5–20-m depth amplitude)
behind the ship in 2001 and horizontally (10–20-m depth) in 2006 and
2008. In addition to temperature, Chl a, and concentration of dissolved
organic matter (CDOM) fluorescence estimates, the BAT contained an
optical plankton counter (OPC; Mini Optical Plankton Counter, model
OPC-2T, Focal Technologies) that counted and sized zooplankton. The
Author's personal copy
W.C. Kerfoot et al. / Journal of Great Lakes Research 36 (2010) 30–41
OPC utilized a flow-through tunnel with a thin rectangular light beam
(4× 20 mm cross-section) that measured the profile area of each
plankton target, and converted the size into an equivalent spherical
diameter (ESD). The device was towed at around 2.5 m/s (ca. 4–
5 knots). Counts were binned in two different size categories (Bin 1,
0.25–0.50 mm; Bin 2, 0.5–1.0 mm). Densities (counts/m3) were plotted
as a 50-interval mean and high-low range of 10-s-duration counts. On
25 April 2008, south of Muskegon, the BAT with the OPC was towed
horizontally from offshore, through the thermocline region, and into the
nearshore zone to record horizontal patterns for zooplankton densities.
OPC counts in Lake Michigan have been compared with plankton
tows (Liebig et al., 2006; Kerfoot et al., 2008) and in the laboratory
(Liebig et al., 2006). In open, dilute waters counts were reliable,
whereas in nearshore turbid waters prior to 2002, they tended to be
higher, by virtue of non-living particles being counted along with
living plankton (Liebig et al., 2006).
In April 2006–2008, vertical zooplankton tows were taken at 24
stations along the E–W transect with a 1-m diameter Puget Sound
plankton net (125 µm Nitex). In April 2007, they were obtained along the
diagonal transect out of Muskegon harbor. The vertical plankton tows
were subsampled with a Hensen–Stempel pipette and counted under a
50× Olympus SZ30 dissection scope equipped with a 2× auxiliary
adapter. The samples allowed calculation of species densities and
indicated changes in community structure from 1998–2001 (EEGLE
Project) to 2006–2008. Veliger larvae of quagga mussels also were
counted in the 2006–2008 samples, although use of 125 µm mesh meant
that smaller sizes were lost. Statistical comparisons utilized Systat
(Wilkinson, 2000).
33
Results
Remote sensing evidence for late-winter phytoplankton blooms
SeaWiFS imagery revealed spatially complex suspended sediment
and Chl a patterns in several Laurentian Great Lakes during winter
(January to April). The spatial patterns were first detected during the
low ice-cover El Niño winter of 1998, largely because SeaWiFS
imagery became available and the lakes were relatively free of ice. A
SeaWiFS image from March 24, 1998, illustrates elevated Chl a (OC2)
late-winter plumes in all ice-free lakes, but most pronounced in
southern Lake Michigan (“doughnut”), southern Lake Huron and
western Lake Erie (Fig. 2). In Lake Superior, elevated Chl a regions
were found off Duluth–Superior Harbors, Chequamegon Bay, and
along the southern shallow coastal shelf off the Keweenaw Peninsula.
Ice-covered regions are indicated in gray. In the lower lakes, the
blooms were even better developed in shallow coastal regions and
were sometimes mixed with drifting ice (Green Bay; Saginaw Bay and
off Port Huron, Lake Huron; western and middle basins, Lake Erie).
The coastal and offshore patterns indicated a previously undetected
relationship between storm-induced gyre formation, coastal water
plus resuspended sediment capture, and late-winter productivity.
The cold-water blooms were not restricted to abnormally warm,
nearly ice-free winters, as images from moderate ice-cover years of
2000–2002 also showed very strong suspended sediment (Rrs 555) and
Chl a (OC2) bloom patterns in late March to early May (Fig. 3, Table 2).
Again, cloud or ice-covered regions (high albedo) are indicated in gray
in the images. In Lake Michigan, ship surveys showed that the patterns
Fig. 2. Chlorophyll a patterns (SeaWiFS imagery) for the Great Lakes on 24 March 1998, an El Niño year. During this winter, ice development (grey area) was exceptionally scarce. Under
prevailing westerly winds, note the movement of Chl a-rich waters out of Duluth/Superior Harbor, Chaquamegon Bay, and Ontonagon River in Lake Superior; rotation of Chl a-rich waters
into a circular ring (‘doughnut’) in southern Lake Michigan; movement of Chl a-rich waters out of Saginaw Bay and circulation around Port Huron into the southern basin of Lake Huron;
intense development in the shallow southwestern basin of Lake Erie off the Maumee River, movement of waters along the northern rim into the middle basin of Lake Erie.
Author's personal copy
34
W.C. Kerfoot et al. / Journal of Great Lakes Research 36 (2010) 30–41
Fig. 3. Late-winter TSM and Chl a patterns, 2000–2002 (SeaWiFS imagery). Left column is total suspended matter (TSM), derived from SeaWiFS 555 nm band reflectance, whereas
right column is Chl a, from OC2 equation (see Methods and materials). Scale bar is given above each column. Dates are provided for the various years.
Author's personal copy
W.C. Kerfoot et al. / Journal of Great Lakes Research 36 (2010) 30–41
Table 2
Using SeaWiFS imagery, we noted the number of days that winter algal blooms were
seen in southern Lake Michigan, Saginaw Bay in Lake Huron, and western Lake Erie.
With SeaWiFS imagery, there were two overpasses per day. The table suggests that latewinter algal blooms were relatively common in the lower Great Lakes and in lower Lake
Michigan before quagga mussels began increasing. Daily cloud cover obscured many of
the scenes in January and February, accounting for low early season totals, whereas
clear skies were more common in March and April, allowing better detection of algal
blooms.
Year
1998
1999
2000
2001
2002
Lake Michigan
Lake Erie
Lake Huron
(Saginaw Bay)
J
F
M
A
J
F
M
A
J
F
M
A
–
–
–
1
–
–
1
–
1
5
5
11
7
5
5
4
5
12
11
5
2
1
1
–
3
–
3
2
2
10
5
12
11
8
7
9
13
8
15
9
–
–
–
–
–
–
1
4
–
2
6
9
10
2
4
7
13
12
10
11
were not image artifacts, as transects verified vertical patterns for Chl a,
temperature, and other variables (colored dissolved organic matter,
transmissivity) associated with the spatially complex, rotating surface
gyre structure (Budd et al., 2002; Kerfoot et al., 2008).
The precise circulation patterns appeared to be basin-specific, subject
to ice-off dates, bathymetry configuration, and prevailing wind stress
and propagated currents (Beletsky et al., 1999). For example, multiple
images showed that in southern Lake Huron, following ice-off, sediments
moved out of Saginaw Bay counterclockwise (cyclonic) around the
southern basin. They also rotated counterclockwise off Port Huron,
delivering sediments and phytoplankton into deeper waters. In Lake
Erie, again following ice-off, Maumee River discharges and resuspended
sediments from the shallow western basin combined and moved
northward and eastward (anticyclonic) into the central and eastern
basins (Figs. 2 and 3). Algal growth was most pronounced in Saginaw
Bay and in the western and middle basins of Lake Erie, as the latter
achieved surface Chl a concentrations as high as 5–15 µg/L. Circulation of
sediments and algae was also evident in southern Lake Michigan,
moving off the eastern coastline near the convergence zone (Figs. 1–3).
Algal activity within the circular offshore ring (‘doughnut’) in Lake
Michigan appeared more diffuse than in Saginaw Bay or Lake Erie, and
reached maximum Chl a concentrations of 2–3 µg /L during 1998–2002.
The important point here is that these phytoplankton blooms were latewinter phenomena (Table 2), distinct from the traditional warmer water
spring blooms that developed behind the coastal ‘thermal bar’ in late
April to early June, and that peaked in May–early June.
Details of ‘doughnut’ in Lake Michigan before quagga mussels
In Lake Michigan, we studied late-winter bloom rings (‘doughnut’)
in much greater detail than in the other lakes. In southern Lake
Michigan, water circulation appeared highly episodic, because it was
strongly wind driven. Wind direction during winter was predominantly from the north to northwest, propagating two counter-rotating
gyres: a counterclockwise-rotating (cyclonic, south basin) gyre to the
right of the wind and a clockwise-rotating (anticyclonic, north basin)
gyre to the left of the wind (Fig. 1; Beletsky and Schwab, 2001;
Beletsky et al., 2003; Schwab and Beletsky, 2003). After a large storm,
the gyres were separated by a convergence zone (Figs. 1–3) along the
downwind shore (Grand Haven to Muskegon) with resulting offshore
flow and a divergence zone along the upwind shore with onshore
flow. Offshore flow in the convergence zone could be seen to extend
into the rotating offshore gyre (Figs. 2 and 3). Turbid water captured
in the convergence zone contained very fine-grained material (grain
size b20 µm), in concentrations between 1 and 5 mg/L, which could
stay in suspension for weeks. Within about 2–7 days after capture, the
nutrient-rich water created a doughnut-shaped algal bloom in water
depths roughly between 40 and 100 m. A subsequent strong storm
35
could create a new nearshore turbid (resuspended sediment) region,
or in the absence of a storm, the offshore ‘doughnut’ ring could close
and dissipate.
How long has the ‘doughnut’ late-winter bloom been occurring?
Remote sensing studies (AVHRR, SeaWiFS) in the Laurentian Great
Lakes are hindered during winter by cloud cover, particularly in
January–February (Stroud et al., 2009; also see Table 2). Yet inspection
of AVHRR imagery confirmed TSM and temperature spatial patterns
from 1992 to the present (16 years). Similar scrutiny of SeaWiFS
imagery revealed winter bloom structures from 1998 until the present, a
continuous 10-year interval. April sampling by GLNPO (Great Lakes
National Program Office) from 1982 until the present (26 years)
indicates strong horizontal patterns for Chl a along our E–W transect
region that are very consistent with the ‘hole’ and circular ‘doughnut’
bands seen in SeaWiFS images (Glenn Warren, unpubl. data). That is,
values are lowest in the ‘doughnut hole’, higher in the left turbid limb,
and highest in the right turbid limb, where PO4 loading from rivers is
known to be greatest (St. Joseph, Kalamazoo, Grand, and Muskegon
Rivers, Kreiss et al., 2004; Pauer et al., 2007).
The ‘doughnut’ structure can be present in January, but is most
pronounced in March–April (Table 2). It can also persist through
multiple storms. For example, in 2000, the circular bloom feature was
present from March 7 until April 25, whereas in 2001, it extended
from April 8 until April 24. If storms are not strong, or near the end of
April, the structure would dissipate as rings expand to fill in the
central ‘hole’. On 19–22 April 01 the ‘doughnut’ was well formed and
the cruise path cut across the hole of the structure, whereas on 12–15
April 06 and 21–23 April, 08, the ring pattern was more diffuse.
Phytoplankton in net tows and water samples from 2001 to 2008
dates documented that the species composition in the doughnut had
large-bodied diatom species [Aulacosira (Melosira), Asterionella,
Tabellaria, Fragilaria; Kerfoot et al., 2008], in addition to open-water
small centric diatoms (Cyclotella), cryptophytes, and flagellates
typically found in deep waters (Bergmann et al., 2004). Deep net
tows were often clogged with Aulacosira. However, zooplankton
samples from 2006 included substantial concentrations of quagga
mussel veliger stages (Fig. 4a), although counts were low-biased
because we used 125 µm netting. However, dramatic differences have
now appeared in the late-winter and early spring spatial distribution
of Chl a patterns. These transformations appear due to the spread of
settled, filtering quagga mussels.
Disappearing ‘doughnut’ in Lake Michigan
Both remote sensing images and ship-board measurements record a
major reduction in Chl a and a dramatic increase in water transparency
since 2002, coincident with increase in dreissenid mussels. Winter
veliger larvae from D. rostriformis bugensis were detected during the
2006 cruise (Fig. 4a). These larvae were not present during the EEGLE
late-winter pelagic collections (1998–2001). In 2006, the veligers were
shown to be differentially concentrated in the rings of the ‘doughnut’,
benefiting from the late-winter productivity pulse (Kerfoot et al., 2008).
During the last three years, veligers were most abundant in 2006, and
declined in 2008 (Table 3).
Whereas zebra mussels (D. polymorpha) have a preference for hard
substrates, the rounded and lighter shell and longer, more vertical
siphon of D. rostriformis bugensis allows this species to colonize softer
substrates. Individual D. rostriformis bugensis were capable of colonizing
substrate under the doughnut, just by settling out of suspension. Over
the interval from 2002 until 2006, the density of settled mussels
increased exponentially (Nalepa et al., 2009). By 2006–2008, mean
densities of attached mussels ranged between 6 and 19 × 103
individuals/m2 at 16–30 m depth, 12 and 13 × 103 individuals/m2 at
30-50 m depth, and 5 × 103 individuals/m2 at 51–90 m depth (Nalepa
et al., 2009, 2010). As the ability of settled adults to clear the water
column is related to the volume of water over the benthic interface, one
Author's personal copy
36
W.C. Kerfoot et al. / Journal of Great Lakes Research 36 (2010) 30–41
Fig. 4. Dreissenid mussels in Lake Michigan waters: A) veliger stage from April southern
transect sampling; B) depth distribution of settled quagga mussels (D. rostriformis
bugensis) in southern Lake Michigan (after Nalepa et al., 2009).
would expect filtration effects from settled adults to be most
pronounced around the shallow rim of the doughnut (Fig. 4b; modified
from Nalepa et al., 2009). The clearing in nearshore waters around the
outer, shallow rim of the “doughnut” is evident in plots of SeaWiFS Chl a
values through time (Fig. 5). Values from the outer edges of the
“doughnut” show pronounced, highly significant (OW 1F10 = 48.0;
p = 4.05E−05; OE 1F10 = 18.0; p = 0.002) declines, whereas values
from inner edges show only a modest, nonsignificant decrease (IW
1F10 = 1.6; p = 0.228; IE 1F10 = 1.6; p = 0.233).
Ship-board studies complemented the remote sensing investigation.
Vertical Seabird CTD casts [64 stations at 1.85 km (i.e., 1 nautical mile)
intervals] and horizontal BAT tows recorded water column differences
across the southern basin on a longitudinal transect, between 86.320°W
Table 3
Mean densities (number/m3) of zooplankton along cross-lake transect, S. Lake
Michigan, comparing 2006 (13–14 April 06), 2007 (2–3 April 07), and 2008 (21–23
April 08) values. Densities are means ±95% confidence limits (n = 9, 2006; n = 7, 2007;
n = 11, 2008); total nauplii are shown separately. Significant (p b 0.05; t-test) declines
between 2006 and 2008 are marked by an asterisk (*). Although veliger densities are
high, these are conservative estimates, because smaller individuals passed through the
120 µm mesh.
Taxa
2006
2007
2008
Nauplii
Diaptomus
*Cyclopoids
*Epischura
*Limnocalanus
Dreissena
8142 ± 2309
3208 ± 950
223 ± 111
750 ± 445
364 ± 198
645 ± 381
5721 ± 1776
2807 ± 704
16 ± 19
1682 ± 2468
85 ± 59
123 ± 48
6016 ± 1360
2912 ± 775
14 ± 8
55 ± 29
30 ± 12
448 ± 219
Fig. 5. Reduction of Chl a around the outer edge of the ‘doughnut’, estimated from
SeaWiFS images: A) Chl a values along southern Lake Michigan transect in 1998, 2006,
and 2009, taken from SeaWiFS images (mean of 4 nearby pixels per site). Numbers on
top of x-axis indicate longitudinal position of the “inner” and “outer” edge regression
values plotted in (B). b) Linear regressions are fit to yearly Chl a values in outer edge
versus inner edge of rings. Paired linear regressions represent east and west limbs of the
‘doughnut’, contrasting reduction in outer (solid symbol) versus inner (hollow symbol)
margin. Regression equations are given for each line (*highly significant slope, p b 0.01).
and 87.745°W. Ship-board vertical CTD profiles for Chl a and water
transmissivity during 2001, 2006, and 2008 across the southern transect
also show pronounced clearing through time, especially along the
shallow edges (Fig. 6; compare 2001 with 2006 and 2008). Both Chl a
and water transmissivity change over the time interval, and shallow
waters become less turbid. Mean CTD temperatures, depth-averaged
Chl a, and mean water transmissivity values emphasize how values have
changed (Fig. 7). Standard deviations for vertical (CTD) site profiles
quantify the statistical variation of values. During the cruises, water
temperature ranged between 2.8 and 4 °C in offshore waters, and above
4 °C behind the thermal bar in coastal waters. Note that the standard
deviations around the mean temperature values are very narrow,
underscoring the vertical mixing of the winter water column in the
rotating gyre. Values for different years document highly significant
temporal shifts toward increased water transparency and decreased Chl
a along the outer edge of the eastern and western “doughnut” rings from
2006 to 2008.
In 2001, before quagga mussels, water transparency (transmissivity) varied between 74 and 85% at deep-water sites and was 80 and
82% in the western and eastern ‘doughnut’ rings. In 2001, Chl a ranged
Author's personal copy
W.C. Kerfoot et al. / Journal of Great Lakes Research 36 (2010) 30–41
37
Fig. 6. Vertical clearing of water column, based on ship-board measurements. Measurements come from Seabird CTD casts taken along the southern transect (see Fig. 1) between
South Haven, Michigan, and Waukegan, Wisconsin. Compare the Chl a and % light transmission values for 2001, 2006, and 2008.
between 1.1–2.6 μg/L and 1.8–2.6 μg/L in the western and eastern
doughnut rings, respectively. After quagga mussel appearance, water
transparency increased progressively to 89% by 2006 and 94–96% by
2008. Chlorophyll a concentrations declined to 0.5–1.7 μg/L by 2006
and 0.4–1.5 μg/L by 2008. The reduction of Chl a between 2001 and
2008 was 56–78% in the western limb and 74–75% in the eastern limb.
Not only did transparency increase and Chl a fall across all
sampling stations, the spatial turbidity pattern changed dramatically
(Figs. 6 and 7). The width of the “doughnut” narrowed 26% and
transparency at mid-coastal regions (40–80 m depth) reversed from
more turbid to less turbid (from 80–82% to 97% transmissivity). Some
of the individual values are now extraordinary for Lake Michigan.
Along the western shoreline in 2008, water transmissivity reached
97% and Chl a dropped to 0.4 μg/L, approaching water clarity and Chl a
concentrations reported for Lake Superior.
Individual vertical CTD profiles for 2001 and 2008 document the
dramatic decline in Chl a concentration at the shallower coastal
margins of the transect, as compared to the central region, although
the decline is present at all stations (Fig. 8). When the % reduction is
plotted versus longitude along the southern transect, the greater
reduction along shallow edges is evident (Fig. 9). Reduction of
chlorophyll related to water depth and a corresponding transparency
pattern suggests progressive filtration effects from settled quagga
mussels. The observed results correspond closely to modeled quagga
filtration effects expected from a combination of density and water
column depth (Vanderploeg et al., 2010).
Zooplankton changes
Some zooplankton changes are also apparent from the crosstransect vertical tows (Table 3). From 2006 to 2008, nauplii and total
herbivorous diaptomid copepods decreased in density (26% and 9%,
respectively). Cyclopoid copepods were dominant in 1998–1999
(Vanderploeg et al., 2007), yet by 2008 they were severely depressed
over 2006 values (94% decline), and often missing from individual
vertical tows. There was also a recent reduction of large-bodied,
omnivorous calanoid copepod species from 2006 to 2008 (Limnoca-
lanus, 93% decline; Epischura, 92% decline). The recent reduction in
over-wintering omnivorous (predatory) species may signal reduced
energy flow up to higher trophic levels, indicating repercussions of
the winter Chl a reduction. The Limnocalanus reduction from 2006 to
2008 reverses an increasing trend from 2003 to 2006 discussed by
Barbiero et al. (2009). Reduced phytoplankton density may begin to
have an impact on veliger densities, although the decline from 2006 to
2008 is not significant.
Comparison of OPC coastal counts also showed some differences.
OPC counts were broken down into the two different-sized bins (Bin
1, 0.25–0.5 mm; Bin 2, 0.5–1.0 mm) used previously (Kerfoot et al.,
2008). The smallest size category was comprised mainly of rotifers,
nauplii, and immature copepodites, whereas the larger was primarily
calanoid copepodites. In 2001 OPC transects, Bin 2 coastal densities
ranged between 4000 and 5000 counts/m3 (mean ± SD = 4.8 ±
0.4 × 103/counts/m3) whereas 2008 transects ranged between 2000
and 4000 counts/m3 (mean ± SD = 3.0 ± 1.9 × 103/counts/m3), significantly lower than the earlier counts (p b 0.0001, t-test on 2001 vs
2008 data sets, N = 1169, t = 206). The significant count reduction
occurs in the same region where there were serious reductions in Chl
a due to quagga activity, again suggesting an impact up the pelagic
food-web. This test is region-specific and hence more powerful than
the vertical zooplankton tows, which are averaged across the entire
transect.
In 2008, across the thermal bar off Muskegon, the smaller size
category was 4-fold higher in nearshore waters (Fig. 10, Table 4) than
in offshore waters. The second size category (0.5–1.0 mm) was
elevated in deep offshore and declined in nearshore waters, although
values peaked near the thermal bar (4 °C). Despite the lower Bin 2
values in the coastal regions, the relative Bin values were similar to
previous historical comparisons, i.e. rotifers were much more
abundant (2–4×) in nearshore regions (Vanderploeg et al., 2007).
Discussion
Remote sensing (SeaWiFS, MODIS) imagery from 1998 to 2003
verified spatially complex suspended sediment and Chl a patterns in
Author's personal copy
38
W.C. Kerfoot et al. / Journal of Great Lakes Research 36 (2010) 30–41
Fig. 7. Statistical plots for depth-averaged CTD values along southern Lake Michigan transect: (A) temperature, (B) Chl a, and (C) % light transmission. Values are means and standard
deviations for vertical casts. Note the severe decrease in Chl a concentration in both the eastern and western shelf regions (down to 0.4 μg/L), the extreme increase in water
transparency (average % transmission 93–98%), and the narrowing of the high-Chl a ‘doughnut’ region.
the Laurentian Great Lakes during late winter. These patterns
suggested an important coupling between late-winter storm-induced
gyre formation, coastal water plus sediment capture, and deep-water
productivity (Kerfoot et al., 2008). Recent resuspended particle-based
modeling efforts have successfully reproduced the resuspension and
transport dynamics seen on SeaWiFS imagery for Lake Michigan (Lee
et al., 2007).
Detailed inspection of river discharges and resuspended sediments
along the southeastern coastline of Lake Michigan (Biddanda and Cotner,
2002) suggested high levels of dissolved organic matter and enhancement of productivity along the coastal shelf in years when the nearcoastal turbidity plume did not hinder photosynthesis (Millie et al., 2002;
Lohrenz et al., 2004, 2008). Here we extended the initial Lake Michigan
observations to a number of the Laurentian Great Lakes. The remote
sensing images revealed an important coastal-offshore linkage, one that
normally benefited over-wintering planktonic species characteristic of
the large lakes. In Lake Erie, the productivity pulse occurred with scattered
ice and persisted into the time of the spring bloom, so that the two
phenomena might be confused. High densities of late-winter algae
probably contributed to the summer deoxygenation of hypolimnetic
waters in the middle (summer ‘Dead Zone’) and eastern basins.
The magnitude of the offshore transfer of nutrients in Lake Michigan
might be exaggerated by climate change, as higher temperatures and
more frequent winter storms suppressed coastal ice formation and
encouraged movement of nutrient-enriched river waters and sediments
into pelagic waters (Kerfoot et al., 2008). The dominant algae associated
with the Lake Michigan ‘doughnut’ ring blooms (Aulacoseira islandica) is
the same species that recently clogged net tows during 2007–8 Lake Erie
ice-breaker studies (Wilhelm, 2008). In Lake Erie, Aulacoseira islandica
was found attached under ice and in open waters near ice, achieving
near-bloom conditions (Mike McKay, unpubl. data).
The late-winter blooms were distinct from traditional spring
blooms, because they occurred earlier and in cold water, away from
the coastal thermal bar. Thermal bar formation and stratification, in
fact, disturbs the conditions that promote offshore transfer of river
waters and resuspended sediments. During low ice years, phosphorus-rich plumes (river water and coastal sediments) are transported
offshore and go through a heterotrophic to autotrophic transformation between January and April in deep waters (Cotner et al., 2000;
Biddanda and Cotner, 2002; Kerfoot et al., 2008). The remote sensing
and ship studies have stimulated preliminary modeling reconstructions (Ji et al., 2002; Beletsky et al., 2003; Lee et al., 2007).
Author's personal copy
W.C. Kerfoot et al. / Journal of Great Lakes Research 36 (2010) 30–41
39
Fig. 8. Comparative depletion of Chl a in shallow shelf versus deep-water, central regions, showing the vertical impact. Vertical profiles for April 2001 (before quagga mussels) are
compared with profiles from April 2008 (after quagga mussels) from western shelf, central, and eastern shelf regions.
Earlier optical plankton counter (OPC) transects and vertical
plankton net tows in Lake Michigan documented that the lateral latewinter transfers influenced higher levels of food webs (zooplankton).
Deep-water, over-wintering zooplankton species (Leptodiaptomus
sicilis, Limnocalanus macrurus, Eurytemora affinis, Senecella calanoides)
seemed to benefit from the late-winter production, as they differentially
congregated in the rings (Kerfoot et al., 2008). However, the discovery in
2006 that veliger larvae from quagga mussels were also exploiting the
late-winter production pulse in southern Lake Michigan, and presumably in the other lower Great Lakes, carried disturbing implications
that are now playing out. Deep-water veligers experience a double
benefit, relative to spread and growth. Not only do the suspended
larvae capitalize on the late-winter productivity pulse, they also
disperse in strong winter currents associated with the winter storms.
Increased benthic attached mussel abundance in coastal waters behind
the thermal bar seems to be severely depleting the spring bloom
(Fahnenstiel et al., 2010; Nalepa et al., 2009, 2010), yet we show here
that it is also reducing the winter ‘doughnut’ pulse in contiguous
offshore waters. As suspended material is filtered from overlying waters,
Fig. 9. Spatial pattern of Chl a depletion along southern transect in Lake Michigan, expressed
as % change (i.e. reduction) between April 2001 sites (before quagga mussels) and April 2008
sites (after quagga mussels). The difference between the mean Chl a site values is plotted
against longitude, along the southern ship transect. The Chl a reduction occurs most
dramatically along the shallower rim (“edges”) on both sides, is symmetrical, and is less in the
deep central basin. The fit polynomial regression (Y=−1.127e06 − 25874X− 148.55X2) has
an r2 =0.901; but more important, it shows the depth-dependent symmetry of the severe
coastal reductions relative to quagga mussel abundance (see Fig. 4).
Fig. 10. Optical plankton counter (OPC) transect from nearshore into offshore waters off
Muskegon, Michigan, illustrating differences in zooplankton densities on 14 April 2008.
Densities are plotted as a 50-interval mean and high-low range for 10-s-duration interval
counts. Total counts (red) and separate densities from two separate size bins (blue, Bin 1,
0.25–0.5 mm; green, Bin 2, 0.5–1.0 mm) are plotted. Dark lines show the mean density (all
three bins) for the three intervals: nearshore (left portion), thermal bar region (middle),
and offshore region (right). Bin 2 densities increase slightly in the region of the thermal bar
(4 °C). Statistics are listed in Table 4.
Author's personal copy
40
W.C. Kerfoot et al. / Journal of Great Lakes Research 36 (2010) 30–41
Table 4
Grand means and standard deviations for BAT optical plankton counter tallies,
nearshore into offshore waters, 14 April 2008, off Muskegon, MI. Grand means are
based on a series of 50-interval means for number of individuals counted during a 10-sduration interval. Estimates of fluid volume flows are then used to convert counts into
density estimates given below.
Location
1.0–7.6 km, nearshore
Total density (×103/m3)
Bin 1 (0.25–0.5 mm)
Bin 2 (0.5–1.0 mm)
8.6–16.4 km, thermal bar
Total density (×103/m3)
Bin 1 (0.25–0.5 mm)
Bin 2 (0.5–1.0 mm)
16.5–37.9 km, offshore
Total density (×103/m3)
Bin 1 (0.25–0.5 mm)
Bin 2 (0.5–1.0 mm)
Mean
Std dev
26.2
24.2
2.4
7.9
7.4
1.5
13.0
7.3
5.4
6.7
4.7
3.0
5.6
2.5
3.0
3.0
1.5
1.9
remarkable water clarity is achieved along with diminished phytoplankton abundance and productivity. The filtering attached mussels
seem to be constricting the winter bloom primarily along the edges and
diminishing even the deep-water algae.
Along the coastal regions of Lake Michigan, water transmissivity
and Chl a values are now approaching Lake Superior values. Primary
production estimates in 1998 ranged from 0.18 to 0.48 g C/m2/d
compared to 0.18–0.24 g C/m2/d for satellite-derived values, and in
2000 ship-board studies gave 0.19–0.37 g C/m2/d compared to 0.15–
0.46 g C/m2/d for satellite-derived values (Lohrenz et al., 2004, 2008).
Recent offshore Muskegon site estimates suggest a 66% decline in
spring Chl a and a 70% reduction in productivity (Fahnenstiel et al.,
2010), whereas GLNPO April ship surveys show about a 72% decrease
in April Chl a concentrations in southern Lake Michigan (Mida et al.,
2010). Both these estimates are similar to our observed declines,
although we document the spatial pattern across southern Lake
Michigan in much more detail.
Quagga mussel filtration is influencing both nearshore and offshore
productivity. Barbiero et al. (2009) argued that a combination of topdown and bottom-up effects between 1998 and 2006 led to a large
increase in Limnocalanus. We found that since 2006, filtration from
settled mussels seems to be diminishing Chl a concentrations for deepwater, over-wintering omnivorous zooplankton species (cyclopoid
copepods; the calanoid copepods L. macrurus and Epischura lacustris).
The OPC counts in sensitive regions over the coastal quagga beds
document a highly significant reduction in counts. We interpret this as a
reduction in zooplankton density for medium sized species and size
classes (Bin 2, 0.5–1.0 mm). However one must be cautious that the
reduced counts might also be correlated with inert particles and
increased water transparency, since OPC counts in Lake Michigan can
include particles other than live plankton (Liebig et al., 2006). The
offshore depression of cyclopoid copepods may be linked to severe
coastal suppression, as absolute densities of cyclopoid copepods were
previously greatest in nearshore waters (Vanderploeg et al., 2007). Latewinter veliger larvae benefit from the winter production pulses because
the late-winter blooms provide a primary source of food at a critical time
when starvation could be important. The late-winter phytoplankton
pulse allows deep-water recruitment to begin in April, before the spring
bloom (Nalepa et al., 2010). In our samples, there are some indications of
a plateau, or even a modest reduction in veliger density, perhaps
suggesting the beginning of a late-winter density-dependent feedback.
That is, diminished algae in the ‘doughnut’ may lower food levels for the
April deep-water pulse of veligers, reducing future settlement success.
Similar density-dependent feedbacks have been recorded for zebra
mussels in Lake St. Clair, Saginaw Bay, Lake Erie, and Lake Champlain
(Bentil et al., 2007).
Larger zooplankton (0.5–1.0 mm) seems decreased over coastal
waters where quagga mussels are most abundant. Increased smaller
zooplankton (especially rotifers) in nearshore waters is expected, and
the ratio of small to large size categories appears as skewed in our OPC
counts as before. For example, during March–April 2000 EEGLE
sampling at four transects (New Buffalo, Gary, Chicago, Muskegon),
relative densities of microplankton and small zooplankton to large
zooplankton increased from deep to nearshore waters, along a ratio of
1.2–2.3× (Vanderploeg et al., 2007).
The implications of quagga-mediated productivity displacements
in the Great Lakes are disturbing. In all probability, in Lake Michigan
we are witnessing a major shift of productivity patterns in time and
space largely mediated by a single species (i.e. quagga mussels). The
spatial and temporal changes resemble shifts documented in Lake Erie
for dreissenid filtering (Depew et al., 2006). As quagga and zebra
mussels filter the overlying water column, their feces and pseudofeces
contribute nutrients and organic matter to the benthic environment
(Vanderploeg et al., 2002). The increased water transparency allows
light to penetrate to the shelf bottom and the displaced nutrients
encourage diatom and attached macrophyte growth (e.g. Cladophora).
This process has been termed ‘benthification’ by Mayer et al. (2002)
and the ‘shunt’ by Hecky et al. (2004).
The LND (‘lateral nutrient displacement’) observed in past latewinter SeaWiFS images should act opposite to ‘benthification’ effects,
differentially transporting shallow-water river discharges and resuspended nutrients back out to the deep pelagic zone. Yet descending
coastal shoals of quagga mussels appear to have interrupted
resuspension of littoral nutrient-rich sediments and lateral transport
of organic-rich river discharges. Our observations of greatly enhanced
water transparency and severely decreased Chl a concentrations
along inner (shallow) basin margins suggests serious interference
with lateral transport. If the spring bloom collapses along the entire
shoreline of Lake Michigan, can the remaining deep-water portion of
the late-winter productivity pulse and the summer deep-water Chl a
maximum adequately sustain the pelagic deep-water food-web?
The presence of dreissenids in the other Great Lakes has caused
dramatic increases in water clarity, increased nuisance benthic plant
growth, and increased harmful algal blooms, with indications of
impacts on many fisheries (e.g. Vanderploeg et al., 2002; Hecky et al.,
2004). We stress that the winter–spring transition is also a critical
time for calanoid copepod reproduction. Deep-water calanoid
copepods should be the most resilient to nearshore coastal impacts.
However, any effect of cold-water dreissenid pelagic veligers or
settled adults on ‘doughnut’ phytoplankton and microzooplankton in
Lake Michigan would influence survival and subsequent reproduction
of over-wintering calanoid copepods during this period, at a time
when food is limiting for over-wintering plankton species, creating a
winter bottleneck. Severe reduction of calanoid and cyclopoid
copepodite stages would also have serious implications for fish
recruitment in the spring, since young of many species draw from
these over-wintering planktonic stages as their primary food source.
Preliminary evidence on a few pelagic Lake Michigan fishes suggests
developing impacts (e.g., alewives; Pothoven and Madenjian, 2008).
Acknowledgments
Research was supported by NSF OCE-9726680 EEGLE CoOP (National
Science Foundation and National Oceanic and Atmospheric Administration Coastal Ocean Program, Episodic Events-Great Lakes Experiment) grant to W.C. K. and J.W. B; a NOAA Coastal Ocean program grant
(NOAA portion of NSF/NOAA EEGLE CoOP Project) to D.J. S. and H.A. V.;
an NSF DEB-0083731 (Biocomplexity) grant to W.C. K.; and a NOAA
Michigan Seagrant Award NA16RG1145, Project R/ER-19 to Kerfoot,
Budd, Green, Vanderploeg, and Schwab. We thank Chris Roussi for the
help with the BAT circuitry, Qili Hu and Aaron Hemme for the sampling
support, and Jennifer Reed for obtaining additional ship support funds.
Author's personal copy
W.C. Kerfoot et al. / Journal of Great Lakes Research 36 (2010) 30–41
Lucille Zelazny aided figure preparation. We also acknowledge the
crews of the RV Laurentian, who guided us successfully across Lake
Michigan during a hazardous time of the year.
References
Austin, J.A., Colman, S.M., 2007. Lake Superior summer water temperatures are
increasing more rapidly than regional air temperatures: a positive ice-albedo
feedback. Geophys. Res. Lett. 34, L06604. doi:10.1029/2006GL029021.
Barbiero, R.P., Bunnell, D.B., Rockwell, D.C., Tuchman, M.L., 2009. Recent increases in the
large glacial-relict calanoid Limnocalanus macrurus in Lake Michigan. J. Great Lakes
Res. 35, 285–292.
Beletsky, D., Saylor, J.H., Schwab, D.J., 1999. Mean circulation in the Great Lakes. J. Great
Lakes Res. 25, 78–93.
Beletsky, D., Schwab, D.J., 2001. Modeling circulation and thermal structure in
Lake Michigan: annual cycle and interannual variability. J. Geophys. Res. 106,
19745–19771.
Beletsky, D., Schwab, D.J., Roebber, P.J., McCormick, M.J., Miller, G.S., Saylor, J.H., 2003.
Modeling wind-driven circulation during the March 1998 sediment resuspension
event in Lake Michigan. J. Geophys. Res. 108 (C2), 3038. doi:10.1029/2001JC001159.
Bentil, D.E., Osei, B.M., Ellingwood, C.D., Hoffman, J.P., 2007. Analysis of a Schnute
postulate-based unified growth mode for model selection in evolutionary
computations. Biosystems 90, 467–474.
Bergmann, T., Fahnenstiel, G., Lohrenz, S., Millie, D., Schofield, O., 2004. Impacts of a
recurrent resuspension event and variable phytoplankton community composition on
remote sensing reflectance. J. Geophys. Res. 109, C10515. doi:10.1029/2002JC001575.
Biddanda, B.A., Cotner, J.B., 2002. Love handles in aquatic ecosystems: the role of
dissolved organic carbon drawdown, resuspended sediments, and terrigenous
inputs in the carbon balance of Lake Michigan. Ecosystems 5, 431–445.
Budd, J.W., Kerfoot, W.C., Green, S., Julius, M., 2002. News. Ocean Color Spectrum 33–34
Spring/Summer.
Budd, J.W., Warrington, D.S., 2004. Satellite-based sediment and Chlorophyll a
estimates for Lake Superior. J. Great Lakes Res. 30, 459–466.
Claxton, W.T., Mackie, G.L., 1998. Seasonal and depth variations in gametogenesis and
spawning of Dreissena polymorpha and Dreissena bugensis in eastern Lake Erie. Can.
J. Zool. 76, 2010–2019.
Cotner, J.B., Johengen, T.H., Biddanda, B.A., 2000. Intense winter heterotrophic
production stimulated by benthic resuspension. Limnol. Oceanogr. 45, 1672–1676.
Depew, D.C., Guildford, S.J., Smith, R.E.H., 2006. Nearshore-offshore comparison of
chlorophyll a and phytoplankton production in the dreissenid-colonized eastern
basin of Lake Erie. Can. J. Fish. Aquat. Sci. 63, 1115–1129.
Eadie, B.J., Schwab, D.J., Leshkevich, G.L., Johengen, T.H., Assel, R.A., Morehead, R., Robbins, J.
A., VanHoof, P.L., 1996. Anatomy of a recurrent episodic event: a winter-spring plume
in southern Lake Michigan. EOS Trans. Am. Geophys. Union 77, 337–338.
Fahnenstiel, G., Pothoven, S., Vanderploeg, H., Klarer, D., Nalepa, T., Scavia, D., 2010.
Recent changes in primary production and phytoplankton in the offshore region of
southeastern Lake Michigan. J. Great Lakes Res. 36 (Supplement 3), 20–29.
Gregg, W.W., 1992. Analysis of orbit selection for SeaWiFS: ascending vs descending
node. NASA Tech. Memo 104566. Vol. 2, 16 pp.
Hecky, R.E., Smith, R.E.H., Burton, D.R., Guildford, S.J., Taylor, W.D., Charlton, M.N.,
Howell, T., 2004. The nearshore phosphorus shunt: a consequence of ecosystem
engineering by dreissenids in the Laurentian Great Lakes. Can. J. Fish. Aquat. Sci. 61,
1285–1293.
Ji, R., Chen, C., Budd, J.W., Schwab, D.J., Beletsky, D., Fahnenstiel, G.L., Johengen, T.H.,
Vanderploeg, H., Eadie, B., Cotner, J., Gardner, W., Bundy, M., 2002. Influences of
suspended sediments on the ecosystem in Lake Michigan: a 3-D coupled biophysical modeling experiment. Ecol. Model. 152, 169–190.
Kerfoot, W.C., Budd, J.W., Eadie, B.J., Vanderploeg, H.A., Agy, M., 2004. Winter storms:
sequential sediment traps record Daphnia ephippial production, resuspension and
sediment interactions. Limnol. Oceanogr. 49, 1365–1381.
Kerfoot, W.C., Budd, J.W., Green, S.A., Cotner, J.B., Biddanda, B.A., Schwab, D.J.,
Vanderploeg, H.A., 2008. Doughnut in the desert: late-winter production pulse in
southern Lake Michigan. Limnol. Oceanogr. 53, 589–604.
Kreiss, R., et al., 2004. Lake Michigan mass balance. Lake Michigan Monitoring
Coordination Council, Muskegon, MI.
Liebig, J.R., Vanderploeg, H.A., Ruberg, S.A., 2006. Factors affecting the performance of
the optical plankton counter in large lakes: Insights from Lake Michigan and
laboratory studies. J. Geophys. Res. 111, C05S02. doi:10.1029/2005JC003087.
Lee, C., Schwab, D.J., Beletsky, D., Stroud, J., Lesht, B., 2007. Numerical modeling of mixed
sediment resuspension, transport, and deposition during the March 1998 episodic events
in southern Lake Michigan. J. Geophys. Res. 112, C02018. doi:10.1029/2005JC003419.
Leshkevich, G.L., Schwab, D.J., Muhr, G.C., 1993. Satellite environmental monitoring
of the Great Lakes: A review of NOAA's Great Lakes Coast Watch Program.
Photogrammetr. Eng. Remot. Sens. 59, 371–379.
Lohrenz, S.E., Fahnenstiel, G.L., Millie, D.F., Schofield, O.M.E., Johengen, T., Bergmann, T.,
2004. Spring phytoplankton photosynthesis, growth, and primary production and
relationships to a recurrent coastal sediment plume and river inputs in southeastern
Lake Michigan. J. Geophys. Res. 109, C10S14. doi:10.1029/2004JC002383.
Lohrenz, S.E., Fahnenstiel, G.L., Schofield, O., Millie, D.F., 2008. Coastal sediment
dynamics and river discharge as key factors influencing coastal ecosystem
productivity in southeastern Lake Michigan. Oceanography 21, 61–69.
41
Mayer, C.M., Keats, R.A., Mills, E.L., Rudstam, L.G., 2002. Zebra mussels as ecosystem
engineers: scale-dependent effects on benthic invertebrates in a large eutrophic
lake. J. North Am. Benth. Soc. 21, 616–633.
McClain, C.R., Cleave, M.L., Feldman, G.C., Gregg, W.W., Hooker, S.B., Kuring, N., 1998.
Science quality SeaWiFS data for global biosphere research. Sea Technology Reprint.
Mida, J.L., Scavia, D., Fahnenstiel, G.L., Pothoven, S.A., Vanderploeg, H.A., Dolan, D.M.,
2010. Long-term and recent changes in southern Lake Michigan water quality with
implications for present trophic status. J. Great Lakes Res. 36 (Supplement 3), 42–49.
Millie, D., Fahnenstiel, G., Carrick, H.J., Lohrenz, S.E., Schofield, O.M.E., 2002. Phytoplankton
pigments in coastal Lake Michigan: Distributions during the spring isothermal period
and relation with episodic sediment resuspension. J. Phycol. 38, 1–11.
Nalepa, T.F., Hartson, D.J., Fanslow, D.L., Lang, G.A., Lozano, S.J., 1998. Declines in benthic
macroinvertebrate populations in southern Lake Michigan, 1980–1993. Can. J. Fish.
Aquat. Sci. 55, 2402–2413.
Nalepa, T.F., Fanslow, D.L., Lang, G.A., 2009. Transformation of the offshore benthic
community in Lake Michigan: recent shift from the native amphipod Diporeia spp.
to the invasive mussel Dreissena rostriformis bugensis. Freshw. Biol. 54, 466–479.
Nalepa, T.F., Fanslow, D.L., Pothoven, S.A., 2010. Recent changes in density, biomass,
recruitment, size structure, and nutritional state of Dreissena populations in
southern Lake Michigan. J. Great Lakes Res. 36 (Supplement 3), 5–19.
O'Reilly, J., Maritorena, S., Mitchell, B., Siegel, D., Carder, K., Garver, S., Kahru, M., McClain, C.,
1998. Ocean color algorithms for SeaWiFS. J. Geophys. Res. 103, 24937–24953.
O'Reilly, J., Maritorena, S., O'Brien, M., Siegel, D., Toole, D., Menzies D., Smith, R., Mueller,
J., Mitchell, B., Kahru, M., Chavez, F.P., Strutton, P., Cota, G., Hooker, S., McClain, C.,
Carder, K., Müller-Karger, F., Harding, L., Magnuson, A., Phinney, D., Moore, G.,
Aiken, J., Arrigo, K., Letelier, R., Culver, M., 2000a. SeaWiFS Postlaunch Technical
Report Series, Volume 11, SeaWiFS Postlaunch Calibration and Validation Analyses,
Part 3, NASA Technical Memorandum.
O'Reilly, J., Maritorena, S., Siegal, D., O'Brien, M., Toole, D., Chavez, F., Strutton, P., Cota, G.,
Hooker, S., McClain, C., Carder, K., Muller-Karger, F., Harding, L., Magnuson, A., Phinney,
D., Moore, G., Aiken, J., Arrigo, K., Letelier, R., Culver, M., 2000b. SeaWiFS postlaunch
calibration and validation analyses, Part 3. In Hooker, S.B. and Firestone, E.R. (eds).
Pauer, J.J., Taunt, K.W., Melendez, W., Kreis Jr., R.G., Anstead, A.M., 2007. Resurrection of
the Lake Michigan Eutrophication Model, MICH1. J. Great Lakes Res. 33, 554–565.
Pothoven, S., Madenjian, C., 2008. Alewife and lake whitefish: changes in consumption
following dreissenid invasions in Lakes Michigan and Huron. N. Am. J. Fish. Manag.
28, 308–320.
Roe, S., MacIsaac, H.J., 1997. Deep-water population structure and reproductive state of
quagga mussels (Dreissena bugensis) in Lake Erie. Can. J. Fish. Aq. Sci. 54, 2428–2433.
Schwab, D.J., Beletsky, D., 2003. Relative effects of wind stress curl, topography, and
stratification on large scale circulation in Lake Michigan. J. Geophys. Res. 108, 3044.
doi:10.1029/2001JC001066.
Schwab, D.J., Eadie, B.J., Assel, R.A., Roebber, P.J., 2006. Climatology of large sediment
resuspension events in southern Lake Michigan. J. Great Lakes Res. 32, 50–62.
Sprung, M., 1993. The other life: an account of present knowledge of the larval phase of
Dreissena polymorpha. In: Nalepa, T.F., Schloesser, D.W. (Eds.), Zebra Mussels:
Biology, Impacts, and Control. CRC Press, Boca Raton, FL, pp. 39–53.
Stroud, J.R., Lesht, B.M., Schwab, D.J., Beletsky, D., Stein, M.L., 2009. Assimilation of
satellite images into a sediment transport model of Lake Michigan. Water Resour.
Res. 45, WO2419. doi:10.1029/2007WR006747.
Stumpf, R.P., Ransibrahmanakul, V., Hughes, K., Sinha, R., Ramachandran, S., Gu, H.,
2000. ESDIM progress report: The evaluation of the atmospheric correction
algorithms for processing SeaWiFS data. Project 392N-SeaWiFS Regional Pathfinder: Reproducing Validation, and Exploitation of SeaWiFS Ocean Color Data.
Stumpf, R.P., 2001. Applications of satellite ocean color sensors for monitoring and
predicting harmful algal blooms. J. Hum. Ecol. Risk Assess. 7, 1363–1368.
Stumpf, R.P., Arnone, R.A., Gould Jr., R.W., Martinolich, P.M., Ransibrahmanakul, V.,
2003. A partially coupled ocean–atmosphere model for retrieval of water-leaving
radiance from SeaWiFS in coastal waters. Algorithm Updates for the Fourth
SeaWiFS Data Reprocessing: SeaWiFS Postlaunch Technical Report Series, vol. 22.
Vanderploeg, H.A., Nalepa, T.F., Jude, D.J., Mills, E.L., Holbeck, K.T., Liebig, J.R., Grigorovich, I.A.,
Ojaveer, H., 2002. Dispersal and emerging ecological impacts of Ponto-Caspian species in
the Laurentian Great Lakes. Can. J. Fish. Aquat. Sci. 59, 1209–1228.
Vanderploeg, H.A., Johengen, T.H., Lavrentyev, P.J., Chen, C., Lang, G.A., Agy, M.A., Bundy,
M.H., Cavaletto, J.F., Eadie, B.J., Liebig, J.R., Miller, G.S., Ruberg, S.A., McCormick, M.J.,
2007. Anatomy of the recurrent coastal sediment plume in Lake Michigan and its
impacts on light climate, nutrients, and plankton. J. Geophys. Res. 112, CO3590.
doi:10.1029/2004JC002379.
Vanderploeg, H.A., Liebig, J.R., Nalepa, T.F., Fahnenstiel, G.L., Pothoven, S.A., 2010.
Dreissena and the disappearance of the spring phytoplankton bloom in Lake
Michigan. J. Great Lakes Res. 36 (Supplement 3), 50–59.
Warrington, D.S., 2001. Great Lakes chlorophyll and turbidity estimates using SeaWiFS
(Sea-viewing Wide Field-Of-View Sensor) imagery. Ms thesis, Michigan Technological University, Houghton, MI. 94 pp.
Wetzel, R.G., Likens, G.E., 2000. Limnological Analyses, 3rd ed. Springer.
Wilhelm, S.W., 2008. Field methods in the study of toxic cyanobacterial blooms: results
and insights from Lake Erie research. In: Hudnell, H.K. (Ed.), Proceedings of the
Interagency, International Symposium on Cyanobacterial Harmful Algal Blooms.
Advances in Experimental Medicine and Biology, pp. 507–519.
Wilkinson, L., 2000. SYSTAT: version 10.0. SPSS.
Wright, D.A., Stetzler-Hamilton, E.M., Magee, J.A., Harvey, H.R., 1996. Laboratory culture
of zebra (Dreissena polymorpha) and quagga (D. bugensis) mussel larvae using
estuarine algae. J. Great Lakes Res. 22, 46–54.